1. Technical Field of the Invention
This invention relates to database searching. More particularly, it relates to building a search request XML response tree from database URLs distributed across a plurality of servers.
2. Background Art
Characteristic of many computer systems and applications known in the art is the distribution of data to many separate databases. These are often distributed among many servers, and a particular user may have data accessible through a plurality of such servers. In the past, when a user desired to search for data, it was necessary to access each server individually.
In the past, search addressing has been bound to application semantics. Such application semantics require that a single instance of an object in a data store be addressed, even when that data store is duplicated for scalability and reliability. There is a need in the art for a more flexible system and method for dynamically addressing data store searches to accommodate scalability and reliability.
Prior art approaches for searching databases in a domain distributed across a plurality of servers include IBM® Lotus® Domino™ Domain search and Domino Extended search. A problem encountered in these search approaches, referred to as the “static URL addressing problem”, results from having tightly coupled server location and object identifier. There is, therefore, a need in the art for a system and method for uncoupling server location and object identifier in order to provide a search access method with improved scalability and reliability characteristics.
Domino, IBM, the IBM Logo, Lotus, Notes, QuickPlace are trademarks of International Business Machines in the United States, other countries, or both.